Diabetes has been referred to as the “undiagnosed epidemic of the third millennium”. Some experts predict the number of diabetics world-wide to triple over the next 15 years to about 320 million. Self-monitoring of blood glucose (SMBG) is considered the quintessential prerequisite for diabetes management and treatment. As will be explained in more detail, most current SMBG systems, whether designed for patient or professional use, still have significant limitations.
The three major types of diabetes are type 1 (formerly insulin-dependent diabetes mellitus, IDDM, juvenile-onset), type 2 (formerly non-insulin-dependent diabetes mellitus, NIDDM, adult-onset), and gestational diabetes. About 130,000 children in the US have type 1 diabetes. Treatment for type 1 consists of insulin injections, diet and exercise.
In type 2 diabetes, treatment may include insulin, but preferably oral glucose lowering agents, diet, weight reduction and exercise. Approximately ninety percent (90%) of diabetics are type 2.
Diabetics are predisposed to heart disease, peripheral vascular disease, stroke, retinopathy, kidney disease and neuropathy. The latter is associated with amputations, silent myocardial infarction and sudden death, and it accounts for over 300,000 hospitalizations in the United States each year. Today's total diabetes-related toll to the US economy (direct and indirect costs combined) is estimated to approach $150 billion.
As a true cure for diabetes remains elusive, tight glucose control will continue to be the sine-qua-non of diabetes combat strategies. The benefits of tight glucose control in curbing diabetes-related complications are now authoritatively documented. This evidence also suggests that a large portion of type 2 diabetics may benefit from tight glucose control and insulin. As worldwide knowledge about diabetes will be nurtured by the information age and media-assisted education, masses of undiagnosed diabetics who would benefit from tight glucose control will eventually be brought into the system. Since testing technology will also further mature, these megatrends will co-functionally establish an enormous market for SMBG in the future.
The mainstay of treatment for type 1 and many type 2 diabetics is SMBG in conceit with responding self-administration of insulin to harmonize glucose levels. Current SMBG systems are typically comprised of a test strip-type, dry chemistry device. The test strip is insertable into a hand-held meter that contains a display that gives the user a read-out of results. Alternately, results can be obtained by comparing reaction colors to printed color charts.
From a provider perspective, the main shortfall is that current systems are generally limited to the measurement of glucose. This is in drastic discord with the concept of diabetes as a multi-factorial metabolic syndrome. From a user point of view, there are still limitations in those features that consumers and users believe to be important, such as (1) minimal invasiveness; (2) speed of analysis and (3) ease of performance and minimal complexity (inconvenience) from primary and auxiliary product mixes.
The majority of presently marketed SMBG systems utilize more or less ‘invasive’ technology (lancing of fingertips) to obtain blood samples in a range between 2 and 30 μL. Non-invasive and minimally invasive technologies have been under active development for years, but made it to market only on a very limited scale due to technical difficulties.
Invasive Systems. Several dry-chemistry technologies exist for testing of whole blood specimens. In most devices, liquid reagents are applied onto solid support substrates by some impregnation or coating method. After solvent evaporation, the dry and therefore stable reagent is contained within a reactive zone or signal member (test field). As the blood sample makes contact with the reagents, a chemical reaction is initiated between analyte in the blood and the reagents on the test field. In most conventional test strips the analyst provides blood to test fields manually, contacting the strip with a drop or unspecified portion of a drop of blood. This technique has limitations with respect to constancy of volume applied and locations on the test field surface contacted by the drop. Consequences can be under- or over-sampling, or heterogeneous distribution of blood and hence reaction signal.
Both photometric and electro-sensimetric detection principles are in use. The vast majority of systems used to employ reflectance photometry, however, in recent years an increasing trend towards electrochemical detection (‘sensors’) has occurred.
In meters that measure reflectance photometry, light of a wavelength absorbed by the colored reaction product is shined onto the surface of the test field and the reflected portion is monitored. In contrast to conventional photometry where absorbance is measured from reduced light transmittance in the direction of the incident beam, reflectance is measured at locations angled away from incident light. As light of varying wavelengths is reflected in different directions, an informed choice must be made as to which incident and reflective angles to select for obtaining a signal that is most sensitively and most specifically related to concentration. Preferably, the photocurrent detector (photodiode) of the metering device is positioned at a location where unspecific scattering is minimal and specific reflectance maximal. However, since the two can usually not be completely spatially separated, pure signals are by definition unobtainable (‘needle in a haystack’ phenomenon). This is one reason why it is so difficult to achieve universal standardization of these systems and why the systems differ so much among each other, resulting in widely scattered method means in proficiency testing surveys.
Another limitation resides in the method by which cellular component of blood is separated from plasma. In older products, plasma was separated by soak through methods into coated bibulous materials or reagent films. Cells were then manually removed from the site of blood application by either washing or wiping, potentially giving rise to significant operator-induced errors. Several newer methods permit separation by means other than washing or wiping. The most frequently used are separation by porous glass fiber fleeces or membranes. In these matrices, pore sizes are chosen so that cellular component is held back within the matrix, whereas plasma diffuses through the separating and into the detection layer.
In most colorimetric test strips the separating layer is sandwiched against the detection layer. The reflectance measurement is then made at the side of the test strip opposite to the side of blood application. To keep needed blood volume low, the thickness of the separation layer is kept at a minimum. An adverse consequence is that spatial separation of red cells from the site of measurement is then so small that the thin zone of separation material that is devoid of cells incompletely shields cells. In instrumented measurements this ‘shining through’ effect of red cells can, as long as the effect is constant, be corrected by calibration or a dual wavelength measurement. However, such corrective methodology makes measurements more complex and less precise.
The shining through effect of red cells is particularly disadvantageous for visual interpretation. It is for this reason that most present-day colorimetric test strips cannot be read visually. Visual interpretation can serve as a confidence check for quantitative results provided by the meter. And in locations where meters are not readily available (rural areas, doctors office, ambulance, third world) concentrations can still be determined semi-quantitatively by visual comparison of reaction colors to standardized color charts. Unfortunately, the feature of visual backup is realized only in a minority of present-day systems.
Non-Invasive (NI) and Semi-Invasive Technology. The goal for the SMBG market, a completely non-invasive glucose monitoring technology, although pursued for over a decade, has so far proven elusive, despite perennial promises from companies in the industry. These failures have led to predictions that completely non-invasive optical technology (infrared or other) may not make it to market in any significant way, for both cost and technical reasons. It is also argued that this lack of success was predictable from early theoretical considerations of signal engineering. These considerations include the numerous and variable challenges of isolating a meaningful signal against a background of overpowering non-specific noise, such as noise from water. An authoritative recent review of NI glucose testing technology concludes that: “ . . . none of the NI experiments reviewed provides proof that the signal is related to actual blood glucose concentration. Clark error grid presentation shows performance that is not acceptable for home glucose meters.”
A promising alternative to non-invasive is “semi-invasive” or minimally invasive testing using interstitial fluid (IF). The only product currently marketed that employs this technology is Glucowatch™ from Cygnus, Inc. It uses electrically stimulated (reverse iontophoresis) glucose extraction from IF into a sensor-equipped sample pad. The product was recently approved by the FDA but only for supplementary (trend) testing. Reported problems with IF sampling are variations in skin thickness and permeability, changes in blood/IF equilibration, sweating, signal instability and skin irritation. Furthermore, the watch must be recalibrated every 12 hours which is done by invasive finger stick measurements.
Several more recent devices employ electrochemical (sensimetric) detection. Good progress in system miniaturization has been achieved with these methods because they can function on whole blood, obviating the need for a plasma-consuming, cell-separating member. In some of these products miniaturization is further aided by provision of capillary sampling techniques. Despite these improvements, a major limitation of sensor methods is that visual backup is completely lost. This places a very heavy burden on the manufacturer as even minor flaws in test strip architecture or signal conductivity could have disastrous consequences. Hematocrit dependence in sensor methods can also be substantial due to ‘dilution’ of the electrochemical reaction milieu by cellular component. Furthermore, in these devices signal output is, as in the case of reflectance measurement, non-linear with respect to concentration, requiring complex mathematics for calibration. Finally, the technical sophistication and manufacturing complexity of the sensor methods makes it difficult to produce them at low cost.
In the future the SMBG market will increasingly be driven by consumer demand, managed care, and cost pressures from third party reimbursement companies. In this environment a market conversion from established and affordable invasive whole blood technology to unproven and costly non-invasive systems appears unlikely. However, it is expected that the market will migrate to invasive systems which minimize invasiveness and its associated pain. As such, the Applicant's minimally invasive and relatively less painful technology is believed by Applicants to better achieve the goals sought by the industry, and be well placed in the direction in which the market is heading.